Risk assessment in determining systemic reactivity to honeybee stings in sting-threatened individuals

We evaluated the diagnostic parameters (sting rate, venom-specific RAST IgE, venom-specific RAST IgG, venom-specific IgG4, venom-specific skin test) of 224 individuals at risk of honeybee sting, assigned them a level of risk on the basis of a hypothetical risk model, and then sting challenged each subject with a live honeybee. Of the 70 subjects at the lowest risk level, only three (4.3%) experienced equivocal or mild systemic reactions, whereas eight (72.7%) of the 11 at the highest risk level experienced systemic reactions, of which five (45.5%) were severe. Increase in risk level was associated with an increased proportion of systemic reactions (p < 0.001). Of the individual diagnostic parameters, venom-specific RAST IgE was the best single predictor of reactivity (likelihood ratio = 0.759, p < 0.01). Venom-specific skin testing at a concentration of 0.1 microgram/ml was a better discriminator than a concentration of 1.0 microgram/ml but did not predict one severe systemic reaction. Three methods of prediction of risk and severity of reaction were examined; the one that assessed risk in the most clinically useful manner was similar to the hypothetical risk model. This model compared well with classical methods of risk assessment and may be used to assess high-risk individuals exposed to honeybees when management could include prophylactic immunotherapy.